Patchwork AI Rules Leave Central Jersey Students in a Fog
Some districts have adopted boilerplate policies with no clear guidance on how students can and can't use apps like ChatGPT.
Sunday, September 14, 2025
*** Editor’s Note: This article contains mentions of suicide. Help is available if you or someone you know is struggling with thoughts of suicide. Please call or text 988, the Suicide & Crisis Lifeline. Every life matters.
Just before the start of the new school year, the parents of 16-year-old Adam Raine sued ChatGPT’s parent company, OpenAI. The parents claim the artificial intelligence chatbot helped their son plan his suicide in April. Court papers say Adam started by asking ChatGPT simple questions. Over months, the complaint says, this grew into planning “a ‘beautiful suicide,’ analyzing the aesthetics of different methods and validating his plans.”
Another similar case involves 14-year-old Sewell Setzer III. His parents sued the makers of Character.AI, calling the chatbot “a deceptive and hypersexualized product” in court documents and accusing parent company Character Technologies and its founders of marketing to children. The lawsuit says Sewell died by suicide in February 2024 after forming an unhealthy relationship with the app.
These tragic cases show why experts say schools must help teens learn skills to handle the dangers of using AI chatbots for emotional support. But an examination by The Central Jerseyan of more than a dozen school districts in the region found exactly what experts feared: AI rules vary widely between districts, with little in common. And not one mention the mental health risks experts say come with teens turning to AI for advice.
In some cases, students living just one town apart can face completely different circumstances. Princeton has a robust AI policy, while Hamilton and Trenton do not have any. In many cases, students are given little clarity on what counts as cheating, collaborating, or being creative.
Recent studies show why clear rules are urgent. A nationwide survey this summer found nearly three out of four U.S. teens have used AI companions. These are chatbots designed to have friendly conversations that feel personal and meaningful. More than half of the surveyed teens said they use AI companions several times a month. A 2024 Stanford study also found teens increasingly turn to AI chatbots for mental health advice.
The survey authors urge schools to create age-appropriate lessons explaining how AI companions are designed to create emotional attachment. They stress teaching students the difference between AI validation and real human feedback. Younger teens are much more likely than older teens to trust advice from AI companions, the survey found.
Leading districts like Shrewsbury and Princeton have highly detailed AI plans. They provide clear guidance for teachers and students on what is considered acceptable use of AI in lessons and schoolwork.
Shrewsbury Superintendent Brent MacConnell called artificial intelligence “the most likely force in the history of education to transform student learning” and emphasized its potential to “level the playing field for accessibility and achievement.” His district’s plan focuses on teacher adoption and uses local expertise, including a parent who works in AI.
Though Shrewsbury was the only district to list specific consequences for violations, Princeton and several other districts referred to separate policies like honor codes for additional guidance.
At the other end of the spectrum, several districts have no policy at all, including Hamilton, Freehold, Lakewood, Woodbridge, and Trenton. In a response to emailed questions, Hamilton Township Superintendent Scott Rocco said his district held back on purpose.
“Implementing a formal policy prematurely could inadvertently limit the creative and educational opportunities that AI presents,” Rocco said. Instead, the district created an AI committee and hosted training sessions for staff.
The Central Jerseyan also emailed questions to superintendents at Freehold, Lakewood, Woodbridge, and Trenton, but received no response.
Many of the examined school districts chose to adopt a policy template recommended by the New Jersey Department of Education. Formally known as Policy 2365, it provides a menu of suggested rules and guidelines. These range from teacher disclosure and approved platforms to citation rules and parent permission. The template also encourages districts to decide which grade levels are permitted to use AI and how.
However, even after adopting the state template, few school districts provide details on how they are implementing the list of recommended policies. Five districts — Edison, Lacey, Montgomery, Toms River, and Westfield — adopted identical policies that simply say, “The district’s AI Plan will include details addressing the following issues.” They then list the general topics provided by the state template with few details.
For some districts, the state template works more like a placeholder than a finished rulebook. Rahway partially adopted Policy 2365 in July. Superintendent Aleya Shoieb said a number of the recommendations were tabled while the district gathers more feedback from faculty and parents. She called this school year a “year of research.”
Toms River Superintendent Michael Kenny said his district’s approach is “AI as a strategic partner,” in a response to emailed questions. Kenny said the district hired an IT expert to guide staff and made AI the theme of the summer retreat for administrators.
Edison, Lacey, Montgomery, and Westfield school officials did not respond to emailed questions.
Overall, the uneven adoption of Policy 2365 highlights how the template approach can still result in very different levels of clarity for students, teachers, and parents depending on the district.
In a recent blog post after the Raine lawsuit was filed, OpenAI wrote, “Our goal is for our tools to be as helpful as possible to people—and as a part of this, we’re continuing to improve how our models recognize and respond to signs of mental and emotional distress and connect people with care, guided by expert input.”
In the lawsuit against Character.AI’s makers, a federal judge in Florida rejected a motion to dismiss the case. In her ruling, U.S. District Judge Anne Conway said the company and its founders “by releasing Character A.I. to the public, created a foreseeable risk of harm” that they “were in a position to control.”
The AI rules that do exist in Central Jersey schools tend to cover the same basic topics. Many district policies generally ban plagiarism and cheating, and stress that AI tools should be used under teacher supervision. This is because current technology designed to catch AI use in schoolwork is not fully reliable, school officials say.
Most districts also focus on professional development to ensure proper classroom use. Even in districts without formal policies, training has centered on responsible use. Hamilton’s superintendent reported running more than 20 sessions since 2023 on topics like ethics, digital citizenship, and reimagining learning in the AI era.
For students and families, the stakes could not be higher. Teens are already experimenting with AI companions and homework tools much faster than schools can regulate them. Parents wonder what guidance — if any — their children will receive in class. Until clearer standards emerge, Central Jersey families will continue to navigate an education system where AI rules depend largely on their ZIP code.